This proposal is to conduct studies aimed at furthering our understanding of the neurocognitive basis of age-related working memory (WM) decline. We propose to implement new neuroimaging and behavioral methods that take account of potential cognitive and physiological factors that may lead to spurious results. fMRI, a neuroimaging tool that relies on hemodynamics of the cerebral vasculature to localize neural activity, has been used to reveal the neural mechanisms of age-related cognitive changes. On the basis of fMRI data, theories about how brain-aging leads to age differences in WM, a fundamental cognitive process that underlies many higher cognitive functions, have been formulated. At the same time, however, other data have indicated that mechanisms of age-related cerebrovascular change could lead to spurious results of age-related fMRI signal differences. The current proposal is to investigate the neural mechanisms of age-related WM change with fMRI methods that take these changes into account. In 4 sets of studies, I will first test a new method for measuring age-related fMRI signal change by accounting for vasomotor (i.e., flow changes during neural activity) and transit (i.e., arrival time of red blood cells) components of fMRI signal. Evidence bearing on this hypothesis would be important because, to date, no methods have been proposed to account for these age-related blood-flow differences. Second, I will implement event-related fMRI procedures to test the hypothesis that dorsal prefrontal cortex (RFC) regions (that may underlie WM executive functions) are disproportionately age affected, compared to other, more ventral RFC regions (that may underlie WM storage processes). These studies uniquely isolate executive WM processes from storage processes while controlling for possible age differences in the strategies subjects adopt to optimize performance. The third and fourth studies examine the neural basis of relationships between age differences in processing speed and WM. These studies will combine refined estimates of neural activity, advanced fMRI, behavioral and statistical methods to study the physiological, neural and cognitive origins of age-related cognitive change. They will inform knowledge of the neural architecture of memory and its decline, the most common and debilitating symptom of aging. The results will also inform development of therapeutic techniques to ameliorate age-related cognitive disorders such as Alzheimer's disease.